Genetics, fitness, and mercury exposure in Isle Royale deer mice (Peromyscus maniculatus).

摘要

With limited data of the appropriate sort from natural populations, important issues remain unresolved regarding the impact of habitat fragmentation on genetic deterioration and population viability. To gain insight into the genetic and demographic consequences of habitat fragmentation, I conducted an observational study of unmanipulated populations of deer mice (Peromyscus maniculatus ) in a naturally fragmented habitat in the Lake Superior watershed. Data were collected from P. maniculatus populations in the Sibley Peninsula (Ontario, Canada), the Keweenaw Peninsula (Michigan, USA), and Isle Royale National Park (Michigan).; I used randomly amplified polymorphic DNA (RAPD) markers to examine the genetic structure of P. maniculatus populations in the Lake Superior watershed, where populations have been fragmented since the last glaciation (Chapter 1). Mainland populations separated by 750 km of contiguous habitat were genetically indistinct, as were populations separated by 60 km on the main island of Isle Royale. However, Isle Royale and mainland populations, separated by 25 km of water, were genetically distinct. Furthermore, genetic differences were detectable between some islet populations separated by only 50--4000 m of water. In addition, genetic diversity of islet populations was inversely related to isolation and, unexpectedly, to island area. The genetic structure of P. maniculatus populations in portions of the long-fragmented Lake Superior watershed appears to have been affected by habitat fragmentation.; Using capture-recapture techniques, I estimated survival rates and population growth rates in four islet populations based on the capture histories of 351 individuals (Chapter 2). Average population growth rate, variance in growth rate, and standard deviation of the log of population size, SD[log(N)], were positively related with all measures of genetic diversity. Using an information-theoretic approach, models including metrics of genetic diversity improved my ability to predict survival. Analysis of molecular variance revealed a small but significant amount of genetic variation could be attributed to differences between mice that survived their first winter versus those that did not. Inbreeding depression appears to decrease the viability for this set of unmanipulated P. maniculatus populations.; I also investigated mercury (Hg) exposure in P. maniculatus across the terrestrial ecosystem of Isle Royale (Chapter 3). Isle Royale mice showed evidence of Hg exposure, with concentrations greater in kidneys than in livers. Although previous work suggested that Hg levels may be higher inside the Sargent Lake watershed of Isle Royale than outside the watershed, this was not the case for the mice in this study. Although Hg concentrations in P. maniculatus tissues were not remarkably high, concern is warranted because: (1) Isle Royale National park is a protected area in a remote location, (2) any exposure in P. maniculatus represents a path for biomagnification in the terrestrial food web, (3) the source of this mercury remains unidentified, and (4) the accumulation of a toxin could magnify the impact of inbreeding depression, alter selection pressures, and increase the risk of extinction in islet populations.